Scikit-learn Crash Course - Machine Learning Library for Python
video description
I do have to question something though, what's so wrong with the original dataset? I understand the racists concerns, but I think it should serve as a reminder of how silly our history actually is, and how far we've come?
It also highlights another very important topic in machine learning: Just because a feature is in a data set, doesn't mean that it needs to be part of the analytics or modeling.
Moreover, part of AI explainability is to identify what features you were given to work with, and wich ones you decided to chose for your model. Therefore, having a feature like this is important so that we all learn to critically analyze the information we've been given, and help us understand which features should be used for modeling.
I'm commenitng on here, as opposed to elsewhere, because it seems like this was something you found bad - and were very passionate about in the video. My personal view is that it should have been kept in the data set from scikit-learn, those of us who are serious about this stuff know how to -drop columns-, and be fully tranparent in how we built our models - including the features used.
Therefore, there is absolutely nothing to hide.
Anyone who builds a model, tells you how well it perfomrs, but can't explain anything other than draw you a picture of how a nueral network works should never be trusted. AI and Machine Learning is science. Science must be transparent, repeatible and reproducible.
It should also serve as a reminder of history. Those who don't know about history tend to repeat it. We need these reminders from that context.
Date: 2022-03-14
Related videos
Comments and reviews: 9
Rick
This is the way everything should be taught!
I love that you present concepts in a structured and systematic way, speaking slowly and clearly, using as few words as possible...
- starting with the concept and talking through drawing a logical diagram (which is so important for developing abstract thinking in terms of high level concepts, which is how we think when we are experienced in something).
- then writing clean, concise code to implement each part of the concept
- showing plots that directly demonstrate the effects of the entire iteration
Too many tutorials make the mistake of talking too much. A lot of videos also either assume too much or too little about the viewer's knowledge.
This seems to confidently stike the nail on the head!
Thanks!
reply
This is the way everything should be taught!
I love that you present concepts in a structured and systematic way, speaking slowly and clearly, using as few words as possible...
- starting with the concept and talking through drawing a logical diagram (which is so important for developing abstract thinking in terms of high level concepts, which is how we think when we are experienced in something).
- then writing clean, concise code to implement each part of the concept
- showing plots that directly demonstrate the effects of the entire iteration
Too many tutorials make the mistake of talking too much. A lot of videos also either assume too much or too little about the viewer's knowledge.
This seems to confidently stike the nail on the head!
Thanks!
reply
Saptarshi
At 48 minutes the explanation for polynomial scaling was not clear, the plot for standard scaler and polynomial scaler was shown as same. Then what was the improvement?
Further, at 1.09hr, the syntax for python code inside plt.plot, could anybody pls explain??
At 57th minute you told there are way more cases with fraud cases than without fraud cases. Is that correct? Because there are 80000 samples and only 196 fraud cases.
reply
At 48 minutes the explanation for polynomial scaling was not clear, the plot for standard scaler and polynomial scaler was shown as same. Then what was the improvement?
Further, at 1.09hr, the syntax for python code inside plt.plot, could anybody pls explain??
At 57th minute you told there are way more cases with fraud cases than without fraud cases. Is that correct? Because there are 80000 samples and only 196 fraud cases.
reply
Rick
Can I ask you how you are able to draw on the screen? I understand you are probably using a Stylus pen over some touch screen surface, which mirrors your display, but what software are you using for that?
reply
Can I ask you how you are able to draw on the screen? I understand you are probably using a Stylus pen over some touch screen surface, which mirrors your display, but what software are you using for that?
reply
elghark
min 56:56. You said that 196 cases out 80000 means there are a lot more -fraud cases-(class 1) that -non fraud cases -(class 0). Why? Isn't it the contrary?
reply
min 56:56. You said that 196 cases out 80000 means there are a lot more -fraud cases-(class 1) that -non fraud cases -(class 0). Why? Isn't it the contrary?
reply
kh
00:19 i did not underestand why after changing k value from 5 to 1 prediction diagram changed ? knn is a classification algoithm but here it was like a regration
reply
00:19 i did not underestand why after changing k value from 5 to 1 prediction diagram changed ? knn is a classification algoithm but here it was like a regration
reply
rajat
sorry...but i totally lost it from metrics onwards...it was too heavy to understand...did not understand even the purpose of the lecture let alone the code...
reply
sorry...but i totally lost it from metrics onwards...it was too heavy to understand...did not understand even the purpose of the lecture let alone the code...
reply
freecodecamp
Before I use up a ton of time for nothing, I want to know if Scikit-learn is capable of Deep Q learning because that's what I've been trying to do
reply
Before I use up a ton of time for nothing, I want to know if Scikit-learn is capable of Deep Q learning because that's what I've been trying to do
reply
Jos-
For the Titanic example: 76% of the women survived, whereas just 16% of the men survived, that would have been a really good classifier to start with
reply
For the Titanic example: 76% of the women survived, whereas just 16% of the men survived, that would have been a really good classifier to start with
reply
ADI
I loved the -racist algorithm- concern you raised. I guess most of us would have ignored it while drowning in fancy algorithms.
reply
I loved the -racist algorithm- concern you raised. I guess most of us would have ignored it while drowning in fancy algorithms.
reply
Add a review, comment
Other channel videos
